Previous residential studies conducted at the ARC showed that heroin dependent individuals may be rapidly inducted onto buprenorphine without producing clinically significant opiate-withdrawal symptoms, may be maintained on daily or alternate-day buprenorphine dosing schedules, and experience a mild to moderate withdrawal syndrome after abrupt withdrawal of buprenorphine. Results from dose- ranging studies indicated an appropriate dose for use in non-residential maintenance treatment. The purpose of this study was to determine the effectiveness of buprenorphine (8 mg sl daily) in maintaining opiate-dependent individuals over 6 months of non-residential treatment as compared to the prototypic treatment drug methadone (20 mg or 60 mg daily). 162 opiate-dependent subjects were randomly assigned after stratification by age, sex, and the results of a naloxone challenge. This study, the largest clinical trial to date assessing the effectiveness of buprenorphine for the treatment of opiate dependence, found buprenorphine as effective as methadone-60 mg on almost all outcome measures, and both significantly more effective than methadone-20 mg. There was no significant difference among groups in urine samples positive for cocaine.